Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 14614
    Prev
    Next
    INDEX
    Explanations

    code-related operations involving event handling and AJAX requests

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Javascript code snippets

    np_acts-logits-general · gemini-2.0-flash

    JavaScript DOM manipulation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    <sup>
    -0.66
    îtra
    -0.50
     ihr
    -0.47
    zuf
    -0.47
    </h3>
    -0.45
    ·
    -0.44
     her
    -0.43
    čios
    -0.43
     super
    -0.41
     pin
    -0.41
    POSITIVE LOGITS
    UserScript
    1.02
    zzleHttp
    0.94
     الحره
    0.94
     MainAxisSize
    0.93
    ఔ
    0.90
    ImageContext
    0.84
     XMLHttpRequest
    0.84
     <=",
    0.83
     CSRF
    0.82
    ItemBackground
    0.82
    Activations Density 0.120%

    No Known Activations